Summary: Our primary goal is to identify and define the actions of proteins that can be used to repair tissues or structures that have been damaged by trauma, neoplasia, or other pathology. We have identified over a dozen novel proteins. Two of these induce the formation of cartilage, one induces the formation of bone and hematopoietic tissue, one plays a key role in defining the overall pattern of the developing embryo, and three others have unknown functions. One of these proteins, Frzb, is a secreted inhibitor of signaling by the Wnt growth factor oncogenes, and is the prototype of a growing family of related proteins. Key feedback interactions between secreted growth factors belonging to the Wnt and Bone Morphogenetic Protein (BMP) families and secreted antagonists of these proteins were elucidated. Localized expression of BMPs 4 and 7, Wnt-8, the BMP antagonists chordin and noggin, and the Wnt antagonist Frzb were found to establish the morphogenetic field giving rise to the earliest precursors of vertebrate muscle. The relationship of expression pattern of these proteins appears to maintain a dynamic balance between Wnt-8 and BMPs 4 and 7 pathways that controls tissue fate. To examine the interactions of these signals, technologies to enable loss-of-function experiments will be useful if not essential. We are therefore isolating genes encoding the proteins under study from X. tropicalis, a diploid relative of X.laevis that may allow construction of deletions using site-specific DNA repair and more facile application of third-generation antisense technologies. We have also prepared recombinant adenoviruses encoding Frzb and ADMP for both high-level protein expression and evaluation in mammalian models of skeletal repair; the latter experiments are in progress. We have isolated amphibian orthologs of CDMP-1/GDF-5 and a novel calmodulin binding protein and have characterized their biological distributions. Functional studies are in progress. We have also isolated a cDNA encoding a protein of unknown function with sequence similar to that of the murine ky gene, associated with profound kyphosis, and have defined its domain of expression. We are also developing procedures to evaluate the effects of morphogens on signal transduction pathways using novel proteomic techniques. This has required development of a novel technique for introducing epitope tags using Type II restriction endonucleases that will be of general applicability.